report no. 99/51
proposal for revision of
volatility classes in
EN 228 specification in
light of EU fuels
directive
Prepared for the CONCAWE Automotive Emissions Management Group by its
Ad-Hoc Gasoline Volatility Group:
J.S. McArragher (Chairman)
T. Aarnink
R. Bazzani
R.F. Becker
J. Graham
J. Hevesi
P.M. Martinez
X. Wauquier
P. Heinze (Technical Coordinator)
Reproduction permitted with due acknowledgement
 CONCAWE
Brussels
January 1999
I
report no. 99/51
ABSTRACT
CONCAWE has reviewed the current gasoline volatility specifications within EN228
relating to hot weather driveability, i.e. RVP, E70 and Vapour Lock Index (VLI), in
anticipation of changes to volatility characteristics after year 2000, due to the impact
of the new EU Fuels Directive (98/70/EC).
This study utilises the assessment of the hot weather driveability (or Hot Fuel
Handling (HFH) performance of current European vehicles and the trends of current
and year 2000 car populations to determine the volatility requirements of individual
European markets for year 2000. The generation and interpretation of such data is
based upon extensive knowledge in this field accumulated by member companies
over many years.
Along with appropriate consideration of future trends in gasoline composition, the
study leads to the conclusion that for year 2000 summer volatility classes, a VLI
specification is no longer necessary. Data also shows that for other seasons
volatility classes (non-summer time), a VLI specification is also generally no longer
necessary. Only during the transition periods between summer and winter for four
markets, identified as critical, might a VLI be used as an alternative solution to
ensure satisfactory driveability.
This report had been made available to CEN / TC19 / WG21 during their review of
EN 228 and contains total customer satisfaction curves as a means of allowing the
appropriate selection of volatility classes for individual European markets in
accordance with their climatic variation and car populations. Further proposals for
changes to year 2000 volatility, in addition to VLI removal, are included in this
report.
Due to the proposed increase in minimum levels of E70 and E100, in addition to the
introduction of an E150 minimum limit, the new version on EN 228 should improve
the cold weather driveability performance of gasoline vehicles.
KEYWORDS
Gasoline volatility, volatility classes, Vapour Lock Index, EN 228, gasoline
specifications, hot fuel handling, hot weather driveability, EU Fuel Directive, market
satisfaction for hot fuel handling
NOTE
Considerable efforts have been made to assure the accuracy and reliability of the information
contained in this publication. However, neither CONCAWE nor any company participating in
CONCAWE can accept liability for any loss, damage or injury whatsoever resulting from the use
of this information.
This report does not necessarily represent the views of any company participating in CONCAWE.
II
report no. 99/51
SUMMARY
The European Standardisation Committee (CEN) has the task of harmonising the
fuel compositional changes, as defined in the EU Fuels Directive for 2000
(98/70/EC), within a new European Standard for 2000. Accordingly, CEN is revising
the year 2000 EN 228 gasoline specifications, which were established for the first
time as a European standard in 1993.
CONCAWE has reviewed the volatility specifications related to hot weather
driveability, i.e. RVP, E70 and Vapour Lock Index, in anticipation of changes to
volatility characteristics after year 2000 due to the impact of the new EU Fuels
Directive. In particular, restrictions on maximum content of olefins, aromatics and
benzene will require changes in refinery processing and a need for increased use of
lower boiling blending components, which would be constrained by the current
volatility specifications after year 2000. An example of this is the parameter, E70.
In view of this, CONCAWE has calculated volatility levels which will ensure
customer satisfaction of hot weather driveability in individual European countries
during their different seasons. By specifying adequate volatility specifications, the
occurrence of hot weather driveability will be avoided and satisfactory performance
ensured.
CONCAWE’s predictions of customer satisfaction for hot fuel handling are based on
a database containing many hundreds of vehicles tested over many years
representing a wide selection of engine technology (carburetted, multi-point injection
(MPI) and single-point injection (SPI). These include an assessment of vehicles’
sensitivity to a wide range of fuel volatility at ambient temperatures representative of
the European market. Using vehicle population data for individual markets and
average monthly maximum temperatures of any individual month of that market
(equivalent to around 95%ile highest temperature), technical satisfaction levels are
established to calculate lines of total customer satisfaction for year 2000. These
total customer satisfaction lines were generated for 14 European markets, for the
most relevant months, as a means of allowing the appropriate selection of volatility
classes.
The analysis of the hot weather driveability (or Hot Fuel Handling (HFH))
performance of the year 2000 car population of different European markets leads to
the conclusion that for year 2000 summer volatility classes, a VLI specification is no
longer necessary. Also for other seasons volatility classes (non-summer time) a VLI
specification is also generally no longer necessary. However, the transition periods,
between summer and winter, of four markets were identified as critical, Finland,
France, Greece and Portugal. Here alternative solutions, such as a VLI for those
months, are required to also ensure that total customer satisfaction is maintained.
The study also demonstrates that the maximum E70 limit can be raised slightly to
48% v/v for summer classes and to 50% for other seasons, without compromising
hot driveability performance.
This will ease some constraints of gasoline
manufacturing which result from restrictions on year 2000 maximum content of
olefins, aromatics and benzene. CONCAWE proposes also to increase the
minimum RVP specifications for the revised volatility classes to reflect concerns
over fuel tank vapour space flammability during cold spells. An increase in E100
maximum is proposed, reflecting the need to compensate for compositional changes
induced by the EU Fuels Directive.
III
report no. 99/51
Cold starting and cold weather driveability performance of vehicles were also
reviewed in view of the proposed modifications, and were found not to be
significantly affected by RVP, but controlled by higher distillation points. It was
therefore concluded that cold weather driveability performance should improve in
year 2000, due to the increase in E70 and E100, together with the introduction of an
E150 minimum limit, in the new version on EN 228.
CONCAWE has made available their proposed volatility classes with detailed
technical background documentation to the experts of Working Group 21 of CEN /
TC19 to serve as guidelines for the respective revisions of EN 228.
IV
report no. 99/51
CONTENTS
Page
SUMMARY
III
1.
INTRODUCTION
1
2.
BACKGROUND TO HOT FUEL HANDLING (HFH)
2
3.
CURRENT FUEL SPECIFICATIONS AND FUTURE DEVELOPMENTS
3
4.
SUMMER VOLATILITY SPECIFICATIONS
4.1.
BACKGROUND INVESTIGATION
4.2.
ANALYSIS
4
4
4
5.
OTHER SEASONS VOLATILITY SPECIFICATIONS
6
6.
TRENDS IN HOT WEATHER DRIVEABILITY PERFORMANCE
7
7.
EFFECTS OF REVISED VOLATILITY CLASSES ON COLD STARTING
AND DRIVEABILITY PERFORMANCE
9
8.
CONCLUSIONS AND PROPOSED SPECIFICATIONS
11
9.
REFERENCES
13
APPENDIX 1
SUMMER AND WINTER MARKET SATISFACTION FOR KEY
MONTHS
14
APPENDIX 2
BACKGROUND TO SATISFACTION ANALYSIS
29
APPENDIX 3
TRENDS IN HOT WEATHER DRIVEABILITY PERFORMANCE
38
V
report no. 99/51
1.
INTRODUCTION
Gasoline volatility specifications in CEN were originally set on the basis of a report
CR 262 “Volatility of Petrol” which concluded that limits should be set for RVP
min/max., E70 min/max., E100 min/max., E180 min. and FBP max.. It also stated
that hot fuel handling performance should be controlled by limits to Vapour Lock
Index (VLI) where VLI = RVP(kPa) x 10 + 7 x E70. After long debate, a series of
eight volatility classes were agreed for the EN 228 specification with limits for RVP,
E70 and VLI, with CEN member countries allowed to choose up to 3 classes to
cover their climatic variation.
The new EU Fuels Directive for 2000 (98/70/EC) specifies RVP at a much lower
level of 60 kPa max. for the summer period, beginning no later than 1 May and
ending not before 30 September. For markets with arctic conditions, RVP will be
limited to 70 kPa max. for a summer period beginning no later than 1 June and
ending not before 31 August (minimum 3 months). This means that some of the
existing volatility classes must be changed to come into line with the new EU limits,
and as such, allows an opportunity to re-examine volatility limits for both summer
and winter.
While the report focuses on hot fuel handling (or hot weather driveability)
performance of European vehicle fleets to propose revised volatility classes in EN
228 specification for year 2000, it also reviews possible effects of such revisions on
cold starting and cold weather driveability.
1
report no. 99/51
2.
BACKGROUND TO HOT FUEL HANDLING (HFH)
If there is a mis-match between the maximum ambient temperature in which a
vehicle is expected to operate and the fuel volatility it uses, then Hot Fuel Handling
(or hot weather driveability) malfunctions can be experienced. These problems are
caused by overheating in the vehicle fuel system leading to vaporisation of the
gasoline. This can cause problems with fuel pumps and metering systems (injectors
or carburettors) which are designed to handle liquid fuel and cannot cope with
vapour. The problems which affect fuel systems can be categorised as:
Carburettor Percolation: Fuel boiling in the float-bowl during a hot-soak forces
excess fuel through the jet into engine, which is hard to restart due to an over-rich
mixture.
Carburettor Foaming: Superheated fuel boils as it enters the float-bowl, generating
foam in which the float sinks, opening the needle valve and allowing more fuel to
enter. Foam blocks the vent causing the float bowl pressure to rise and forcing
excess fuel through the jet into the engine. The engine will not run due to an overrich mixture.
Fuel Pump/Injector Vapour Lock: Fuel boils in the pump, fuel rail or injectors,
forming slugs of vapour which prevent the carburettor or injectors from metering
liquid fuel. The engine will not start or misfires due to an over-lean mixture.
Modern multi-point electronic fuel injection (EFI) engines are much less prone to all
HFH problems than carburetted engines. This is mainly because of the higher
operating pressure of the fuel system (preventing vaporisation) and re-circulation
systems which serve to cool the injectors and dissipate heat energy into the fuel in
the tank. Therefore modern vehicles are much more tolerant of hot conditions and
high volatility fuels and there are very few cases of HFH problems in the market.
The effect of fuel properties on HFH has been widely studied (see below) and the
key parameters are as follows:
T V/L(X) Temperature to give Vapour/Liquid Ratio (x): x can be 10 to 40, but
typically 20. This correlates well with HFH, and is used in US ASTM specification.
However it is hard to measure, and is generally calculated from a nomogram.
VLI (Vapour Lock Index) or FVI (Flexible Volatility Index) = RVP (kPa) x 10 + 7 x
E70 : This index was developed in the 1970s and correlates with TV/L(X) . It is
generally accepted as the best parameter to describe fuel hot-weather performance
of current vehicles (1)1.
ASVP at 100°C: This has been proposed as an alternative to VLI, but oil industry
tests have shown it gives no better correlation with HFH than VLI.
1
2
numbers in brackets designate references given in section 9
report no. 99/51
3.
CURRENT FUEL SPECIFICATIONS AND FUTURE DEVELOPMENTS
Current CEN Volatility specifications are shown in Table 1.
Table 1
Current CEN EN 288 Volatility Specifications
CLASS
1
RVP hPa
2
3
4
5
6
7
8
350-700 350-700 450-800 450-800 550-900 550-900 600-950 650-1000
E70 %v/v
15-45
15-45
15-45
15-45
15-47
15-47
15-47
20-50
VLI max.
900
950
1000
1050
1100
1150
1200
1250
40-65
40-65
40-65
40-65
43-70
43-70
43-70
43-70
E180 %v/v min
85
85
85
85
85
85
85
85
FBP °C max.
215
215
215
215
215
215
215
215
Residue %v/v
2
2
2
2
2
2
2
2
E100 %v/v
Volatility characteristics will change after year 2000 due to the impact of the new EU
Fuels Directive. In particular, restrictions on maximum content of olefins (18% v/v),
aromatics (42 %v/v) and benzene (1 %v/v) will require changes in refinery
processing. There will need to be increased use of lower boiling blending
components, such as isomerate and MTBE. CONCAWE studies show that because
of these changes to gasoline production, the current limit on maximum E70 (45 47% v/v) will be constraining after year 2000. Therefore the needs of the European
car parcs have been analysed for year 2000, for individual countries. The parc
response to fuel volatility has been determined to predict whether the current E70
and VLI limits are still necessary to maintain problem-free hot fuel handling
performance and to assess if there is scope for relaxation.
3
report no. 99/51
4.
SUMMER VOLATILITY SPECIFICATIONS
4.1.
BACKGROUND INVESTIGATION
In order to fully understand where the current volatility classes within EN 228 are in
relation to driveability satisfaction levels, and to determine future volatility
requirements, a detailed investigation was carried out at one of CONCAWE’s
member companies. The company operates a technical facility that can determine
the level of driveability satisfaction of any western European market, for any fuel
volatility, based on vehicle performance and vehicle population data for that market.
Vehicle performance data are generated through participation in an inter-industry
working group, whose purpose is to generate both hot and cold weather driveability
data on a selection of vehicle technologies representative of the European market at
different ambient temperatures. These vehicles are assessed for their sensitivity to
a wide range of fuels of different volatility. The company’s in-house database
contains many hundreds of vehicles tested over many years. These performance
data when linked, on a market weighted basis, with the vehicle population data and
an accurate ambient temperature profile of a region, allow hot weather driveability
technical satisfaction levels to be generated for any combination of ambient
temperature and volatility. In this analysis, the average monthly maximum
temperature at the hottest location in a market is used to define either the hottest
month in the season under review, e.g., as shown for the summer below, or for
individual months, as specified for other seasons (Figure 2 and Appendix 1). These
technical satisfaction levels are then used to determine levels of market satisfaction
as explained in further detail in Appendix 2.
4.2.
ANALYSIS
This exercise was carried out for 14 European markets: Austria, Benelux, Denmark,
Finland, France, Germany, Greece, Italy, Norway, Portugal, Spain, Sweden,
Switzerland and UK. For each of these markets, the current EN 228 summer
volatility classes were superimposed on the driveability satisfaction plot, at the
highest summer season average monthly maximum temperature. This was carried
out for 1993, “current” 1997 and a projected year 2000 vehicle fleet which is based
on changes in vehicle fleet hot driveability performance extrapolated from current
trends as explained in Appendix 2. An example of a single market, France, is
shown below in Figure 1.
4
report no. 99/51
Figure 1
France - Summer Market Satisfaction for Hot Fuel Handling
Market Satisfaction for Hot Fuel Handling - France
RVP bar
Current EN228:
EU 2000:
Summer (20.06 to 09.09) - 35 ° C } = average maximum temperature of hottest month
Summer (01.05 to 30.09) - 35 ° C }
1.2
Total Customer Satisfaction 1993
1.1
Predicted Total Customer Satisfaction 2000
1.0
Total Current ('97) Customer Satisfaction
0.9
Year 2000
RVP = 60kPa
0.8
0.7
Current EN228 Class 1
0.6
Proposed volatility box
0.5
0.4
0.3
10
15
20
25
30
35
40
45
50
55
E70 %vol
The French car parc contains a significant number of older vehicles which are fitted
with carburetted fuel systems and in combination with high summer temperatures
are more likely to exhibit hot driveability malfunctions referred to above. Figure 1
clearly shows that the volatility class chosen in 1993, Class 1, was closely aligned
with the vehicle requirements, and was thus suitable for the French car parc at that
time. This demonstrates the validity of the customer satisfaction analysis, as further
described in Appendix 2. However as discussed below vehicle hot driveability
performance has improved substantially in recent years due to the progressive
replacement of carburetted vehicles by those fitted with EFI, as clearly shown by the
satisfaction lines for current (1997) and projected year 2000 vehicle fleets. Thus
when RVP is reduced to 60kPa max. in year 2000, the predicted “Total Customer
Satisfaction 2000” line shows that separate limits on VLI will be unnecessary and
there is technical justification to allow an increase in maximum E70 to 48% v/v.
5
report no. 99/51
5.
OTHER SEASONS VOLATILITY SPECIFICATIONS
To investigate the volatility requirements for other seasons, similar calculations have
been carried out for 14 European markets. These calculations are based on hot
driveability performance of the predicted year 2000 vehicle fleet for individual
markets (details are given in Appendix 2). They show predicted year 2000
customer satisfaction levels for average monthly maximum temperatures in
individual months, covering the critical transition months between winter and
summer. On these charts have been superimposed the summer volatility limits and
a suggested class to cover the other seasons, based on current maximum RVP
levels. An example of a critical market, France, is shown in Figure 2, and a full set of
these diagrams for all countries is included in Appendix 1. The proposed volatility
classes are given in Tables 2 and 3.
Figure 2
France - Year 2000 Market Satisfaction for Hot Fuel Handling in
different months
Market Satisfaction for Hot Fuel Handling - France
Jul, Aug - 35°C
Jun - 33°C
Sep - 32°C
May - 29°C
Oct - 27°C
Apr - 26°C
Mar - 24°C
Nov - 22°C
Feb - 21°C
1.2
1.1
RVP bar
1.0
0.9
Class D
0.8
0.7
0.6
Class A
0.5
0.4
10
15
20
25
30
35
40
45
50
55
E70 %vol
It can be seen that for France, the proposed volatility Class D would give adequate
customer satisfaction for months March and November (and all colder months) but
would be marginal in April and October. As the new EU summer period from year
2000 covers May to September for non-arctic countries, there is a possibility that
driveability could be a concern in April and October. This could be overcome either
by retaining an intermediate grade, as is currently the case, or by extending the
summer period slightly. A more detailed study of temperature data during April and
October is needed to determine the best option. If this approach is taken, it is clear
that even for this critical market there is no need for separate VLI limits and an
increase of E70 up to 50% v/v is possible without compromising hot driveability
performance. Appendix 1 shows that this is the case for all markets, and only in
Finland, France, Greece and Portugal is there need for extra control of volatility
levels during transition periods between summer and other seasons.
6
report no. 99/51
TRENDS IN HOT WEATHER DRIVEABILITY PERFORMANCE
As mentioned earlier in this document, modern multi-point electronic fuel injection
(EFI) engines are much less prone to HFH problems than carburetted engines and
are much more tolerant of hot conditions and high volatility fuels. In order to
quantify this impact relative to the overall driveability performance of a market, 2
examples were selected to represent critical and non-critical markets, namely Spain
and Germany, respectively, as shown in Appendix 3. The same hot weather
driveability satisfaction analysis was again applied, but to vehicles registered in a
particular year. By doing this, a driveability performance trend over several years
can be established, as shown below in Figure 3 for the Spanish market.
Figure 3
Spain - New Registration Trends for Hot Weather Driveability
Trends in Total Customer Satisfaction - New Registrations
Spain - Summer
1996 Satisfaction
Off Scale
1982 New Registrations
1986 New Registrations
1990 New Registrations
1992 New Registrations
1996 New Registrations
Total Customer Satisfaction 1996
1.2
1.1
1.0
RVP bar
6.
> 1993 Satisfaction
Off Scale
0.9
0.8
0.7
0.6
0.5
10
15
20
25
30
35
40
45
50
55
E70 %vol
Between 1982-1986, approx. 90-95 % of vehicles sold from new were fitted with
carburetted fuel systems (see Appendix 3) and this is reflected in the relatively poor
driveability performance of the vehicles registered during this period. A major
improvement in vehicle performance is seen by 1990-1992 (which coincides with a
drop from 70 to 50% penetration of newly registered carburetted vehicles in the
market) and by 1993, the improvement becomes so dramatic that the total customer
satisfaction line is off scale in Figure 3. This coincides with the mandatory
introduction of 3-way catalyst vehicles which were predominantly equipped with fuel
injection systems as a technical requirement for greater control of engine
stoichiometry.
Clearly carburetted vehicle fleets have poorer driveability
satisfaction. This is confirmed when comparing the ‘Total Customer Satisfaction
1996’ line which is based on the total 1996 car parc, to the 1990 ‘New Registration’
line. The two lines are virtually identical and indeed the number of fuel injected
vehicles on the road in Spain in 1996 was ca. 30% which compares to the ca. 30%
of vehicles registered in 1990 with fuel injection.
7
report no. 99/51
The data shown suggests that as carburetted fuel systems are replaced with fuel
injection, (EU carburetted population estimated to be ≤ 25% for 2000), markets
within Europe will become even less sensitive to fuel volatility. This change will be
most evident in critical markets which have currently a high proportion of carburetted
vehicles. By way of illustration, the Italian market had been marginal in customer
satisfaction in October using the 1996 car parc data (2). By including the 1997 car
parc data and revised predictions, the Italian market has become less sensitive to
fuel volatility for year 2000 and would not require extra control of volatility levels
during transition any more with the proposed winter classification. Details of the
change in hot weather driveability performance for the Italian market are discussed
in Appendix 2.
8
report no. 99/51
7.
EFFECTS OF REVISED VOLATILITY CLASSES ON COLD
STARTING AND DRIVEABILITY PERFORMANCE
When summer RVP levels were reduced in USA some problems of poor starting
and driveability were reported during the transition periods before and after summer.
It is therefore important to ensure that this will not happen when the revised volatility
classes are introduced in Europe as this could lead to an increase in exhaust
emissions and impairing customer satisfaction. However, for the reasons given
below, such problems are unlikely to emerge in Europe.
In Europe, only RVP will be reduced in the summer period, and hence potentially
during critical transition periods, whereas both E70 and E100 minimum limits will be
increased in ALL countries and ALL months. In practice, minimum (and maximum)
RVP levels will only be reduced in six countries during transition periods, generally
in October, but also in some countries in April or May. The problems in USA were
not due directly to a reduction in RVP, but to reduction in other mid-range and frontend volatility properties which was consequent on the change in RVP.
Cold starting performance of vehicles is not significantly affected by RVP, but is
controlled by higher distillation points. Early fundamental work (3) showed that time
to start is dependent on air/fuel vapour ratio which can be calculated from distillation
properties. This showed that E70 was the controlling property which was confirmed
in earlier (1970) CONCAWE member company work for carburettor vehicles (4).
This work showed that at -18 °C cold starting correlated best with T10E or E70,
while at -7 °C the best correlation was with E100 or E120. More recent work on fuel
injected vehicles (5) has confirmed that for temperatures between 0 and 16 ºC
(more typical of transition periods) the fuel property which correlated best with
starting was E100 or T50, giving significantly better correlation than E70. The cold
start performance of modern vehicles is very good, such that in this recent work,
cold start times greater than 2 seconds were only seen for fuels well outside
proposed specification limits.
Similarly Cold Weather Driveability is not affected by RVP but is correlated with
mid-range volatility i.e. E100 or T50E, as discussed in a recent CONCAWE
document (6). Where driveability has been correlated with other distillation
properties, as in the US Driveability Index, front-end volatility (T10 or E70) has only
shown a relatively weak effect. Recent CRC work (7) suggested that an Index based
on E100 and E150 gives the best correlation, which is supported by another recent
SAE paper (8). Analysis by a CONCAWE company of driveability performance of
modern European cars showed that less than 10% of cars tested showed RVP to
have a significant positive effect. This compared with 50% for E100 and 47% for
E70. Thus CWD performance should be significantly improved by the increase in
E100, in addition to the introduction of an E150 minimum limit, as defined in the year
2000 fuel specification.
There is undoubtedly a correlation between exhaust emissions and driveability, as
driveability malfunctions are a manifestation of engine misfires which will increase
HC and CO emissions. This was clearly shown in the EPEFE programme where
significant increases in emissions together with reported driveability problems were
seen on several of the gasoline vehicle fleet on 2 fuels (EPGA2 and EPGA3), both
of which were significantly below the proposed future minimum E100 limit. Work by
GM (9) has also demonstrated a clear correlation between HC emissions and
driveability. A CONCAWE programme (10) also showed that fuel volatility did affect
exhaust emissions, but the effects were relatively small. It is clear that emissions
9
report no. 99/51
only increase significantly at very low levels of mid-range volatility, well below the
year 2000 European standard.
10
report no. 99/51
8.
CONCLUSIONS AND PROPOSED SPECIFICATIONS
Analysis of the Hot Fuel Handling (HFH) performance of current European vehicles,
the volatility requirements and trends of the current and year 2000 car populations,
and future trends in gasoline composition, leads to the following conclusions:
•
For year 2000 summer volatility classes, a VLI specification is no longer
necessary. The RVP maximum limit of 60 / 70 kPa will already ensure total
customer satisfaction.
•
For year 2000 other season volatility classes a VLI specification is generally
no longer necessary. The proposed volatility levels for other seasons (nonsummer time) will also be for most countries below the requirement to ensure
total customer satisfaction (Appendix 1). Where this is not the case (Finland,
France, Greece and Portugal) alternative solutions are required to also
ensure total customer satisfaction during transition periods.
•
The customer satisfaction analysis demonstrates that the maximum E70 limit
can be raised slightly to 48% v/v for summer classes and to 50% for other
seasons, without compromising hot driveability performance. This would
accommodate the use of lower boiling blending streams to meet lower limits
on olefins, aromatics and benzene. A further review is recommended for the
EU 2005 revision.
•
Minimum RVP specifications have been increased to reflect concerns over
fuel tank vapour space flammability during cold spells. This has been
discussed in CONCAWE report 97/53.
•
E100 and E150 minimum specifications are included to reflect the revisions
agreed in the new EU Fuels Directive. An increase in E100 maximum is also
proposed, reflecting the need to compensate for compositional changes
induced by the EU Fuels Directive. Other volatility specifications are included
for completeness.
•
Cold starting and cold weather driveability performance of vehicles are not
significantly affected by RVP, but are controlled by higher distillation points.
Increased minimum levels of E70 and E100, together with the introduction of
an E150 minimum, will all be introduced in the new version on EN 228, so
cold driveability performance should improve. Emissions only increase
significantly at very low levels of mid-range volatility, well below the year 2000
European standard.
Proposed specifications are given in Tables 2 and 3 below.
Table 2
Proposed new Summer CEN Gasoline Volatility Classes.
CLASS
A
B
RVP kPa
45-60
45-70
E70 %v/v
20-48
20-48
E100 %v/v
46-71
46-71
E150 %v/v min.
75
75
FBP °C max.
215
215
Residue %v/v
2
2
11
report no. 99/51
Table 3
Proposed new CEN Gasoline Volatility Classes for other
Seasons
CLASS
12
C
D
E
F
RVP kPa
50-80
60-90
65-95
70-100
E70 %v/v
22-50
22-50
22-50
22-50
E100 %v/v
46-71
46-71
46-71
46-71
E150 %v/v min.
75
75
75
75
FBP °C max.
215
215
215
215
Residue %v/v
2
2
2
2
report no. 99/51
9.
REFERENCES
1.
CEN (1991) Volatility of petrol. Report No. CR 262, prepared by technical committee
CEN/TC19 Test methods and specification for petroleum products. Brussels: Comité
2.
CONCAWE (1998) Draft working document 'Proposal for revision of volatility
classes in EN 228 specification in light of EU fuels directive', tabled at the
CEN / TC19 / WG21 meeting, 9 December 1998, Brussels. Document No. 98/02.
Brussels: CONCAWE
3.
IP (1946) Modern petroleum technology. London: Institute of Petroleum
4.
Shell (1971) Internal report
5.
Shell (1994) Internal report
6.
CONCAWE (1998) Technical view on gasoline driveability index, September 1998
7.
Jorgensen, S.W. et al (1996) A new CRC cold-start and warm-up driveability test
and associated demerit weighting procedure for MPFI vehicles. SAE Paper
No. 962024. Warrendale: Society of Automotive Engineers
8.
Stephenson, T. and Luebbers, M. (1998) Evaluating the performance of driveability
indices: a correlation with the enthalpy of vapour formation for gasoline. SAE Paper
No. 982722. Warrendale: Society of Automotive Engineers
9.
Jorgensen, S.W. and Benson, J.D. (1996) A correlation between tailpipe
hydrocarbon emissions and driveability. SAE Paper No. 962023. Warrendale:
Society of Automotive Engineers
10.
CONCAWE (1993) The effect of gasoline volatility on vehicle emissions at low
ambient temperatures. Report No. 93/51. Brussels: CONCAWE
11.
Palmer, F.H. (1985) Hot weather driveability: does the CEC CF24 test method
reflect motorists requirements? CEC 2nd International Symposium on The
performance evaluation of automotive fuels and lubricants, Wolfsburg, June 5-7,
1985. Paper EF3.
12.
Becker, R.F. and Tontodonati, A.N. (1985) European hot fuel handling tests with
gasoline fuelled vehicles. SAE Paper No. 852128. Warrendale: Society of
Automotive Engineers
13
report no. 99/51
THIS PAGE HAS BEEN LEFT BLANK INTENTIONALLY
14
report no. 99/51
APPENDIX 1
SUMMER AND WINTER MARKET SATISFACTION
FOR KEY MONTHS
15
report no. 99/51
Market Satisfaction for Hot Fuel Handling - Austria
Year 2000 Satisfaction - Forecast Using 1997 Data
1.2
Jul - 33°C
Aug - 32°C
Jun - 31°C
Sep - 29°C
May - 28°C
Apr - 25°C
Oct - 24°C
1.1
RVP bar
1.0
0.9
Class D
0.8
0.7
0.6
Class A
0.5
0.4
10
16
15
20
25
30
35
E70 %vol
40
45
50
55
report no. 99/51
Market Satisfaction for Hot Fuel Handling - Benelux
Year 2000 Satisfaction - Forecast Using 1997 Data
Jul - 32°C
Aug, Jun - 31°C
Sep - 29°C
May - 28°C
Apr, Oct - 23°C
1.2
1.1
RVP bar
1.0
0.9
0.8
0.7
Class E
0.6
Class A
0.5
0.4
10
15
20
25
30
35
E70 %vol
40
45
50
55
17
report no. 99/51
Market Satisfaction for Hot Fuel Handling - Denmark
Year 2000 Satisfaction - Forecast Using 1997 Data
Jul - 29°C
Aug, Jun - 27°C
May - 25°C
Sep - 23°C
Apr, Oct - 18°C
1.2
1.1
RVP bar
1.0
0.9
0.8
0.7
Class E
0.6
Class A
0.5
0.4
10
18
15
20
25
30
35
E70 %vol
40
45
50
55
report no. 99/51
Market Satisfaction for Hot Fuel Handling - Finland
Year 2000 Satisfaction - Forecast Using 1997 Data
Jul - 29°C
Jun - 27°C
Aug - 26°C
May - 23°C
Sep - 20°C
Apr - 16°C
1.2
1.1
RVP bar
1.0
Class F
0.9
City Gasoline
Winter Specification
0.8
0.7
Class B
0.6
0.5
0.4
10
15
20
25
30
35
E70 %vol
40
45
50
55
19
report no. 99/51
Market Satisfaction for Hot Fuel Handling - France
Year 2000 Satisfaction - Forecast Using 1997 Data
Jul, Aug - 35°C
Jun - 33°C
Sep - 32°C
May - 29°C
Oct - 27°C
Apr - 26°C
Mar - 24°C
Nov - 22°C
Feb - 21°C
1.2
1.1
RVP bar
1.0
0.9
Class D
0.8
0.7
0.6
Class A
0.5
0.4
10
20
15
20
25
30
35
E70 %vol
40
45
50
55
report no. 99/51
Market Satisfaction for Hot Fuel Handling - Germany
Year 2000 Satisfaction - Forecast Using 1997 Data
Jul - 33°C
1.2
Aug - 32°C
1.1
Jun - 31°C
May, Sep - 29°C
RVP bar
1.0
0.9
Class D
0.8
0.7
0.6
Class A
0.5
0.4
10
15
20
25
30
35
E70 %vol
40
45
50
55
21
report no. 99/51
Market Satisfaction for Hot Fuel Handling - Greece
Year 2000 Satisfaction - Forecast Using 1997 Data
1.1
Jul, Aug - 40°C
Jun - 38°C
Sep - 37°C
May, Oct - 33°C
1.0
Apr - 30°C
Nov - 27°C
RVP bar
1.2
0.9
0.8
Class C
0.7
0.6
Class A
0.5
0.4
10
22
15
20
25
30
35
E70 %vol
40
45
50
55
report no. 99/51
Market Satisfaction for Hot Fuel Handling - Italy
Year 2000 Satisfaction - Forecast Using 1997 Data
Jul, Aug - 40°C
Jun - 37°C
Sep - 36°C
May - 32°C
Oct - 30°C
Apr - 28°C
Nov - 27°C
Mar - 24°C
1.2
1.1
RVP bar
1.0
0.9
Class D
0.8
0.7
0.6
Class A
0.5
0.4
10
15
20
25
30
35
E70 %vol
40
45
50
55
23
report no. 99/51
Market Satisfaction for Hot Fuel Handling - Norway
Year 2000 Satisfaction - Forecast Using 1997 Data
July - 28°C
Jun, Aug - 27°C
May - 24°C
Sep - 21°C
Apr - 17°C
Oct - 16°C
1.2
1.1
RVP bar
1.0
Class F
0.9
0.8
0.7
Class B
0.6
0.5
0.4
10
24
15
20
25
30
35
E70 %vol
40
45
50
55
report no. 99/51
Market Satisfaction for Hot Fuel Handling - Portugal
Year 2000 Satisfaction - Forecast Using 1997 Data
Jul - 42°C
Aug - 40°C
Jun, Sep - 38°C
May - 33°C
Oct - 32°C
Apr - 29°C
Mar - 26°C
Nov - 24°C
Feb - 22°C
Dec - 20°C
1.2
1.1
RVP bar
1.0
0.9
0.8
0.7
Class E
0.6
Class A
0.5
0.4
10
15
20
25
30
35
E70 %vol
40
45
50
55
25
report no. 99/51
Market Satisfaction for Hot Fuel Handling - Spain
Year 2000 Satisfaction - Forecast Using 1997 Data
Jul - 43°C
Aug - 42°C
Jun - 39°C
Sep - 38°C
May - 34°C
Oct - 33°C
Apr - 30°C
1.2
1.1
RVP bar
1.0
0.9
0.8
Class C
0.7
0.6
Class A
0.5
0.4
10
26
15
20
25
30
35
E70 %vol
40
45
50
55
report no. 99/51
Market Satisfaction for Hot Fuel Handling - Sweden South
Year 2000 Satisfaction - Forecast Using 1997 Data
Jul - 30°C
1.2
RVP bar
Jun, Aug - 28°C
1.1
May - 25°C
1.0
Sep - 23°C
City Fuel
Winter
Specification
0.9
0.8
Class E
0.7
0.6
0.5
Class B
0.4
10
15
20
25
30
35
E70 %vol
40
45
50
55
27
report no. 99/51
Market Satisfaction for Hot Fuel Handling - Switzerland
Year 2000 Satisfaction - Forecast Using 1997 Data
1.2
Jul - 34°C
1.1
Aug - 33°C
Jun - 32°C
Sep - 30°C
RVP bar
1.0
May - 29°C
Apr - 25°C
0.9
0.8
0.7
Class E
0.6
Class A
0.5
0.4
10
28
15
20
25
30
35
E70 %vol
40
45
50
55
report no. 99/51
Market Satisfaction for Hot Fuel Handling - UK
Year 2000 Satisfaction - Forecast Using 1997 Data
Jul - 29°C
Aug,Jun - 28°C
Sep - 26°C
May - 25°C
Apr - 21°C
Oct - 20°C
1.2
1.1
RVP bar
1.0
Class F
0.9
0.8
0.7
0.6
Class A
0.5
0.4
10
15
20
25
30
35
E70 %vol
40
45
50
55
29
report no. 99/51
THIS PAGE HAS BEEN LEFT BLANK INTENTIONALLY
30
report no. 99/51
APPENDIX 2
BACKGROUND TO SATISFACTION
ANALYSIS
31
report no. 99/51
REQUIREMENTS FOR ESTABLISHING SATISFACTION ANALYSIS
There are three criteria required to analyse a market for hot weather driveability satisfaction.
Firstly, an accurate ambient temperature profile of a market under analysis, which should include
a range of temperature data for each month averaged over many years. Data supplied by the
UK Meteorological Office, London was used in this analysis. From this, the average monthly
maximum temperature is used to approximate the 95% worst case (equates to a temperature
being held for 1 hour in any month) and the hottest month in any defined season, or for individual
months, is used as the criteria.
Secondly, the vehicle population of a market which identifies the exact type and number of
vehicles on the road in any market. This population data is purchased from DRI MacGrawHill,
London on an annual basis.
Thirdly, the driveability performance of these vehicles at different ambient temperatures and fuels
of different volatility. Such data is generated by the Intercompany Volatility Group whose
purpose is to generate hot and cold weather driveability data on the CEC driveability cycle on a
wide selection of vehicle technologies representative of the EU market. These tests are carried
out by trained raters under critical driving conditions and are a severe test of vehicle
performance.
The above data is used to calculate the percentage of vehicles technically satisfied (i.e. with no
driveability problems) for any combination of temperature and fuel volatility. The market
satisfaction analysis employed in this document is derived using the latest car parc data available
which at the time of writing correspond to the 1997 car parc.
TECHNICAL VS. CUSTOMER SATISFACTION
Several chassis dynamometer / road trials were performed during validation of the CEC Hot
Weather Driveability Procedure by the CEC CF-24 Driveability Group. These confirmed that CEC
tests carried out on the dynamometer using trained raters under controlled conditions are more
severe than on the road, but little knowledge was gained in how technical assessments reflected
the actual driveability requirements of the motorist. Thus in order to establish motorist’s
perception to changes in volatility / temperature, a series of summer season trials were carried
out independently by both BP (11) and Shell using common test cars and fuels.
Results from all these trials showed that the motorist and technician response to changes in
volatility were similar (i.e. fuels of higher volatility generally created more driveability
malfunctions) but their levels of perception were different. Motorists were considerably less
critical of driveability malfunctions and fuel volatility than technically trained raters. The main
reasons for this are as follows:
•
Chassis Dynamometer tests give rise to higher fuel system temperatures due to reduced
air flow over the vehicle. CEC tests (12) showed that this difference is equivalent to a
3.4°C change in ambient temperature or 90 mbar change in FVI at constant temperature.
•
Motorists rarely drive in the manner of the CEC test cycle. Approximately 5% of driving
corresponds to the portion of the CEC test cycle that will generate problems.
•
Carburetted vehicles on dynamometer are tested with vapour traps, so fuel will not
“weather” as it will on the road where high volatile fuels will lose some vapour into the
atmosphere.
32
report no. 99/51
The BP trials showed that total customer satisfaction was equivalent to a technical satisfaction
level of 70%. Work by Shell showed that total customer satisfaction was equivalent to a
technical satisfaction of 80% (see Figures 4 and 5). The difference can be explained by the fact
that the Shell analysis corrects test results for the dynamometer-road temperature difference
whereas the BP calculations do not. Thus even though a different method of analysis was used,
it can be seen in Figure 6 that both approaches to determining customer satisfaction are in very
close agreement.
VALIDATION OF CUSTOMER SATISFACTION PREDICTIONS
The preceding paragraphs have discussed specific approaches to the analysis of hot driveability
data generated by vehicle tests under strictly controlled conditions on chassis dynamometers. A
critical step in the process is the interpretation and treatment of this data to predict customer
response to road driveability performance from the response of trained technical raters on the
dynamometer. Figures 4 and 5 show how technical and customer satisfaction levels were
related in limited fleet tests conducted separately by BP and Shell.
The validity of selecting 70% technical satisfaction (BP analysis method) to represent Total
Customer Satisfaction is confirmed on a much larger scale, by examining the hot driveability
requirements of a full market population. Using the French market as an example, the predicted
curves for total customer satisfaction have been calculated for the 1993 vehicle population in
both summer and winter. The 1993 vehicle population was chosen to represent the situation
when CEN EN 228 was introduced. These curves are shown in Figures 7 (summer) and 8
(winter), in comparison with the CEN volatility classes (1 and 6 respectively) chosen by DHYCA
for those periods.
For summer, the curve for 1993 total customer satisfaction is slightly above the VLI line for CEN
Class 1. For winter, the curve for 1993 total customer satisfaction is virtually identical to the VLI
line for CEN Class 6. These diagrams indicate that both the CEN Classes were well matched to
the seasonal volatility requirements of the 1993 French vehicle population. In both cases, the
analysis results suggest that total customer satisfaction was achieved in 1993 and in all
subsequent years (because of the decreasing sensitivity of the population). This analysis has
been borne out in the French market, as CONCAWE is not aware of any customer complaints
that could be related to hot driveability problems and gasoline volatility.
PREDICTION OF HOT WEATHER DRIVEABILITY PERFORMANCE OF YEAR
2000 CAR POPULATION
As previously mentioned, the analysis employed to determine the current level of EU market
satisfaction is derived using 1997 car parc data, which are the latest data available at the time of
writing. As the volatility limits within EN 228 specification will be revised for 2000, it is necessary
to predict the hot weather driveability performance of a projected year 2000 vehicle population.
This was performed for each individual EU market as described below. Essentially the annual
change in hot-weather driveability satisfaction of each country's fleet was determined for each
year between 1993 (mandatory introduction of catalyst equipped vehicles) and 1997. The
average annual change in volatility to achieve customer satisfaction over this period was then
used to extrapolate the change in performance to 1999, i.e. two years. 1999 fleet calculations
have been used to represent the vehicle fleet which will be present in January 2000. This
approach is more focussed on changes in individual markets than the first analysis (2) which was
based on an EU average change in vehicle hot driveability performance.
An example of the predictive analysis for a single market, Italy, is shown in Figure 9. Individual
driveability lines representing total customer satisfaction are depicted from 1993 to 1997
33
report no. 99/51
(inclusive) at a temperature representative of the proposed transition period (April and October)
which, for this market, approximates to 30 °C. The delta RVP, or difference in RVP between the
individual years' satisfaction lines (in mbar) are also shown in Figure 9, averaged over a range of
E70 from 20 to 50 %v/v.
It can be clearly seen that the driveability performance of the Italian market has improved
between 1993 to 1997, as expected, due mainly to the progressive replacement of carburetted
vehicles by those with fuel injection. This improvement increases steadily on a yearly basis as
can be seen by the increase in delta RVP from 9 mbar between 1993 to 1994 to 30 mbar
between 1996 to 1997. On this basis, it could be technically justified to conclude that, for the
Italian market, this trend would continue (assuming a reasonably constant vehicle scrappage
and steady new registrations) over the next 2 years, up to the end of 1999.
However, in order to ensure complete customer satisfaction, a more conservative approach was
taken. It was decided to use the average delta over the years between 1993 to 1997. On this
basis, the annual delta RVP for the Italian market was calculated at 20mbar. This implies that
the predicted 1998 satisfaction line would be 20mbar higher than for 1997, and that the predicted
1999 satisfaction line would be 40 mbar greater than for 1997.
This method of analysis was carried out for all other EU markets at the temperature
corresponding to their proposed transition period. The calculated delta RVP for each market is
shown in Figure 10. For comparison the European average predictions to year 2000 based on
car parc data including 1996 data are shown which have been used in the first analysis (2).
In order to verify this new updated prediction from a driveability satisfaction point of view,
satisfaction lines for the Italian and French market have been compared to the first analysis (2)
as can be seen in Figures 11 and 12. This demonstrates the validity of the revised analysis.
34
report no. 99/51
Figure 4
Technical vs. Customer Satisfaction (11)
35
report no. 99/51
Figure 5
Technical vs. Customer Satisfaction (extracted from Shell Data)
70%
80%
90%
Figure 6
Comparison of BP vs. Shell Technical Satisfaction
HFH Market Satisfaction
UK 1982-92
1600
Critical VLI
1400
1200
1000
800
600
20
60%
36
25
30
35
Ambient Temp (DegC)
70%
80%
40
Shell 80% Technical Satisfaction
report no. 99/51
Figure 7
Total Customer Satisfaction for France (Summer 1993)
Market Satisfaction for Hot Fuel Handling - France
Summer (20.06 to 09.09) - 35°C
Introduction of EN228 1993
1.2
1.1
RVP bar
1.0
0.9
0.8
0.7
EN228 Class 1
VLI = 900 max.
0.6
0.5
0.4
0.3
10
Figure 8
15
20
25
30
35
E70 %vol
40
45
50
55
Total Customer Satisfaction for France (Winter 1993)
Market Satisfaction for Hot Fuel Handling - France
Winter (01.11 to 09.04) - 24°C
Introduction of EN228 1993
1.2
1.1
RVP bar
1.0
0.9
EN228 Class 6
VLI = 1150 max.
0.8
0.7
0.6
0.5
10
15
20
25
30
35
E70 %vol
40
45
50
55
37
report no. 99/51
Figure 9
Total Customer Satisfaction for Italy from Year 1993 to 1997
Market Satisfaction for Hot Fuel Handling - Italy
1993 Total Customer Satisfaction
1.2
1994 Total Customer Satisfaction
1.1
1995 Total Customer Satisfaction
1996 Total Customer Satisfaction
RVP bar
1.0
1997 Total Customer Satisfaction
0.9
0.8
0.7
0.6
0.5
10
15
20
25
30
35
E70 %vol
40
45
50
55
47.5
13
32
10
27
50 52.5
4
4
33
26
18
29
43
35
E70 %v/v
Year
94-93
95-94
96-95
97-96
10
22
11
11
9
12.5
4
11
22
16
15
24
17
11
12
17.5
23
20
9
28
20
21
20
10
26
22.5
24
16
14
27
25
4
8
35
35
27.5
3
5
37
36
30
4
10
28
48
32.5
3
22
34
30
35
6
17
32
20
37.5
9
18
29
25
40
3
43
22
25
42.5
8
19
29
17
45
9
9
18
27
Average
55 (20-50)
4
9
8
19
20
24
82
30
Average change in RVP per year
Figure 10
20
European Average and Individual Market Predictions for Yearly
Change in RVP for Total Customer Satisfaction
Prediction of Yearly Increase in Market Satisfaction for Europe
E s tima tio n B a s e d o n
1 9 9 6 C a r P a rc &
EU Avera g e P re d ic tio n s
Ye a rly D e lta (mbar)
E s tima tio n B a s e d o n
1 9 9 7 C a r P a rc &
Individual Marke t P re d ic tio n s
Ye a rly D e lta (mbar)
Austria
B e n e lux
Denmark
F in la n d
France
Germany
Greece
Ita ly
Norway
Portugal
S p a in
Sweden South
S w itze r la n d
UK
20
20
20
20
20
20
20
20
20
20
20
20
20
20
30
35
33
13
29
41
23
20
20
23
15
24
33
28
Ave ra g e
Min.
Max.
20
-
26
13
41
Country
1
38
report no. 99/51
Figure 11
Year 2000 Predicted Total Customer Satisfaction for Italy
Comparing 1997 and 1996 Calculations
Market Satisfaction for Hot Fuel Handling - Italy
Temperature = 30°C
1.2
RVP bar
1.1
1.0
0.9
1996 Total Customer Satisfaction
0.8
2000 Predicted Total Customer Satisfaction (Using '96 Data)
1997 Total Customer Satisfaction
0.7
1997 Predicted Satisfaction (Using '96 Data)
2000 Predicted Total Customer Satisfaction (Using '97 Data)
0.6
10
15
Figure 12
20
25
30
35
E70 %vol
40
45
50
55
Year 2000 Predicted Total Customer Satisfaction for France
Comparing 1997 and 1996 Calculations
Market Satisfaction for Hot Fuel Handling - France
Temperature = 25°C
1.2
1.1
RVP bar
1.0
0.9
0.8
1996 Total Customer Satisfaction
0.7
2000 Predicted Total Customer Satisfaction (Using '96 Data)
0.6
1997 Total Customer Satisfaction
0.5
1997 Predicted Satisfaction (Using '96 Data)
2000 Predicted Total Customer Satisfaction (Using '97 Data)
0.4
10
15
20
25
30
35
E70 %vol
40
45
50
55
39
report no. 99/51
THIS PAGE HAS BEEN LEFT BLANK INTENTIONALLY
40
report no. 99/51
APPENDIX 3
TRENDS IN HOT WEATHER DRIVEABILITY PERFORMANCE
41
report no. 99/51
Fuel System / Catalyst Penetration among New Registrations
SPAIN
100
%age of new registrations
90
80
70
Single Point Injection
Multi Point Injection
Carburettor
60
50
40
30
20
10
0
80
81
82
83
84
85
86
87
Year
88
89
90
91
92
93
94
95
96
100
90
registrations/Parc
%age of new
80
70
60
Reg
50
Parc
40
30
20
10
0
80
81
82
83
84
85
86
87
88
Year
42
89
90
91
92
93
94
95
96
report no. 99/51
Trends in Total Customer Satisfaction - New Registrations
Germany - Summer
1980 New Registrations
1985 New Registrations
1990 New Registrations
1996 New Registrations
Total Customer Satisfaction 1996
1.2
1.1
1996 New Registrations
Off Scale
RVP bar
1.0
0.9
0.8
0.7
0.6
0.5
10
15
20
25
30
35
E70 %vol
40
45
50
55
43
report no. 99/51
%age of new registrations
Fuel System / Catalyst Penetration among New Registrations
GERMANY
100
90
80
70
60
50
40
30
20
10
0
Single Point Injection
Multi Point Injection
Carburettor
80
81
82
83
84
85
86
87 88
Year
89
90
91
92
93
94
95
96
1 0 0
9 0
8 0
registrations/Parc
%age of new
7 0
6 0
R e g
5 0
P a r c
4 0
3 0
2 0
1 0
0
8 0
44
8 1
8 2
8 3
8 4
8 5
8 6
8 7
8 8 8 9
Y e a r
9 0
9 1
9 2
9 3
9 4
9 5
9 6